Update app.py

app.py

@@ -30,6 +30,22 @@ STANDARD_COLORS = {
     'WORK OF ART': '#DDA0DD'  # Plum
 }
 
+# Entity definitions for glossary
+ENTITY_DEFINITIONS = {
+    'DATE': 'Absolute or relative dates or periods',
+    'EVENT': 'Named hurricanes, battles, wars, sports events, etc.',
+    'FAC': 'Facilities - Buildings, airports, highways, bridges, etc.',
+    'GPE': 'Geopolitical entities - Countries, cities, states',
+    'LANG': 'Any named language',
+    'LOC': 'Non-GPE locations - Mountain ranges, bodies of water',
+    'MISC': 'Miscellaneous entities - Things that don\'t fit elsewhere',
+    'NORP': 'Nationalities or religious or political groups',
+    'ORG': 'Organizations - Companies, agencies, institutions, etc.',
+    'PER': 'People, including fictional characters',
+    'PRODUCT': 'Objects, vehicles, foods, etc. (Not services)',
+    'Work of Art': 'Titles of books, songs, movies, paintings, etc.'
+}
+
 # Additional colours for custom entities
 CUSTOM_COLOR_PALETTE = [
     '#FF9F43', '#10AC84', '#EE5A24', '#0FBC89', '#5F27CD',
@@ -734,19 +750,25 @@ def process_text(text, standard_entities, custom_entities_str, confidence_thresh
 def create_interface():
     with gr.Blocks(title="Hybrid NER + GLiNER Tool", theme=gr.themes.Soft()) as demo:
         gr.Markdown("""
-        # 🎯 Named Entity Recognition (NER) Explorer Tool
+        # Named Entity Recognition (NER) Explorer Tool
         
         Combine common NER categories with your own custom entity types! This tool uses both traditional NER models and GLiNER for comprehensive entity extraction.
-        
-        ## 🤝 Overlapping entities are automatically shared with split-colour highlighting!
-        
+          
         ### How to use:
         1. **📝 Enter your text** in the text area below
         2. **🎯 Select a model** from the dropdown for common entities
         3. **☑️ Select common entities** you want to find (PER, ORG, LOC, etc.)
         4. **✨ Add custom entities** (comma-separated) like "relationships, occupations, skills" - powered by GLiNER
         5. **⚙️ Adjust confidence threshold** 
-        6. **🔍 Click "Analyse Text"** to see results with organized output
+        6. **🔍 Click "Analyse Text"** to see results with organized output 
+                (Common/custom entities which overlap are shown with split-colour highlighting)
+        """)
+        
+        # Add tip box
+        gr.HTML("""
+        <div style="background-color: #fff3cd; border: 1px solid #ffeaa7; border-radius: 8px; padding: 12px; margin: 15px 0;">
+            <strong style="color: #856404;">💡 Top tip:</strong> All models can both miss entities and/or miss categorise entity types - so keep an eye out for this.
+        </div>
         """)
         
         with gr.Row():
@@ -806,7 +828,7 @@ def create_interface():
                 )
             
             with gr.Column():
-                gr.Markdown("### ✨ Custom Entity Types")
+                gr.Markdown("### ✨ Custom Entity Types (Powered by GLiNER)")
                 custom_entities = gr.Textbox(
                     label="Custom Entities (comma-separated)",
                     placeholder="e.g. relationships, occupations, skills, emotions",
@@ -818,8 +840,71 @@ def create_interface():
                 - emotions, actions, objects  
                 - medical conditions, treatments
                 - financial terms, business roles
+                
+                *GLiNER model will extract these custom entity types from your text*
                 """)
         
+        # Add glossary here (Option 1: below selection box but above analyse button)
+        gr.HTML("""
+        <details style="margin: 20px 0; padding: 10px; background-color: #f8f9fa; border-radius: 8px; border: 1px solid #ddd;">
+            <summary style="cursor: pointer; font-weight: bold; padding: 5px; color: #1976d2;">
+                ℹ️ Entity Type Definitions (Click to expand)
+            </summary>
+            <div style="margin-top: 10px; padding: 10px;">
+                <dl style="margin: 0; font-size: 14px;">
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #00B894;">PER:</dt>
+                        <dd style="display: inline; margin-left: 5px;">People, including fictional characters</dd>
+                    </div>
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #55A3FF;">ORG:</dt>
+                        <dd style="display: inline; margin-left: 5px;">Organizations - Companies, agencies, institutions, etc.</dd>
+                    </div>
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #A0E7E5;">LOC:</dt>
+                        <dd style="display: inline; margin-left: 5px;">Non-GPE locations - Mountain ranges, bodies of water</dd>
+                    </div>
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #F9CA24;">GPE:</dt>
+                        <dd style="display: inline; margin-left: 5px;">Geopolitical entities - Countries, cities, states</dd>
+                    </div>
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #45B7D1;">FAC:</dt>
+                        <dd style="display: inline; margin-left: 5px;">Facilities - Buildings, airports, highways, bridges, etc.</dd>
+                    </div>
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #FF6B6B;">DATE:</dt>
+                        <dd style="display: inline; margin-left: 5px;">Absolute or relative dates or periods</dd>
+                    </div>
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #4ECDC4;">EVENT:</dt>
+                        <dd style="display: inline; margin-left: 5px;">Named hurricanes, battles, wars, sports events, etc.</dd>
+                    </div>
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #8E8E93;">NORP:</dt>
+                        <dd style="display: inline; margin-left: 5px;">Nationalities or religious or political groups</dd>
+                    </div>
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #6C5CE7;">LANG:</dt>
+                        <dd style="display: inline; margin-left: 5px;">Any named language</dd>
+                    </div>
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #FD79A8;">MISC:</dt>
+                        <dd style="display: inline; margin-left: 5px;">Miscellaneous entities - Things that don't fit elsewhere</dd>
+                    </div>
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #E17055;">PRODUCT:</dt>
+                        <dd style="display: inline; margin-left: 5px;">Objects, vehicles, foods, etc. (Not services)</dd>
+                    </div>
+                    <div style="margin-bottom: 8px;">
+                        <dt style="font-weight: bold; display: inline; color: #DDA0DD;">Work of Art:</dt>
+                        <dd style="display: inline; margin-left: 5px;">Titles of books, songs, movies, paintings, etc.</dd>
+                    </div>
+                </dl>
+            </div>
+        </details>
+        """)
+        
         analyse_btn = gr.Button("🔍 Analyse Text", variant="primary", size="lg")
         
         # Output sections
@@ -848,7 +933,7 @@ def create_interface():
             outputs=[summary_output, highlighted_output, results_output]
         )
         
-        # Add examples
+        # Add examples (removed the financial terms example as requested)
         gr.Examples(
             examples=[
                 [
@@ -858,13 +943,6 @@ def create_interface():
                     0.3,
                     "entities_spacy_en_core_web_trf"
                 ],
-                [
-                    "The meeting between CEO Jane Doe and the board of directors at Microsoft headquarters in Seattle discussed the Q4 financial results and the new AI strategy for 2024.",
-                    ["PER", "ORG", "LOC", "DATE"],
-                    "corporate roles, business events, financial terms",
-                    0.4,
-                    "entities_flair_ner-ontonotes-large"
-                ],
                 [
                     "Dr. Emily Watson published a research paper on machine learning algorithms at MIT. She collaborates with her colleague Prof. David Chen on natural language processing projects.",
                     ["PER", "ORG", "Work of Art"],
@@ -881,6 +959,7 @@ def create_interface():
                 model_dropdown
             ]
         )
+        
         # Add model information links
         gr.HTML("""
         <hr style="margin-top: 40px; margin-bottom: 20px;">